Authors
Loading...
Analysis shows language models perform similarly to humans on comprehension tasks, suggesting reevaluation of human performance.
Rakshit et al. (2026) studied this question.